Google Analytics Hostname Filters: A Complete GA4 Guide

What Is a Hostname Filter in GA4?

A hostname is simply the domain an event was collected on — example.com or www.example.com, for instance. A hostname filter is a property-level data filter that lets you control which events GA4 actually processes into your reports, based on that hostname. You can either block a handful of known bad domains, or flip it around and only allow data from the domains you trust.

It lives in the same place as two filters you may already know: the internal traffic filter and the developer traffic filter. You'll find all three under Admin → Data collection and modification → Data filters.

Why Did Google Add This to GA4?

Your GA4 Measurement ID is embedded in your site's tracking code, which means it's technically visible to anyone who views your page source. Spammers sometimes copy that ID and send fake events straight into your property, either by cloning it on a lookalike site or firing hits directly at Google's collection endpoint. The result is "ghost" traffic that never actually touched your website but still shows up in your sessions, bounce rate, and conversion numbers — the exact metrics that decisions like a SEO vs. paid ads budget call often rest on.

If you're used to Universal Analytics' referral exclusion list or View-level filters, it's worth noting those don't exist anymore. UA stopped collecting new data on July 1, 2023, and Google deleted all historical UA data by mid-2024. The hostname filter is GA4's native answer to a problem that UA used to solve differently.

Exclude vs. Include Only — Which Filter Type Should You Use?

When you create the filter, you'll choose between two modes. Include filters are generally the safer long-term choice once you know your full hostname list, because you're not stuck chasing every new spam domain that pops up. Exclude filters are fine for a quick, targeted fix.

How Do You Check Which Hostnames Are Sending You Data?

Before touching the filter, confirm what's actually hitting your property. GA4 treats example.com and www.example.com as two separate hostnames, so both need to be listed if you're using exact matching.

• Go to Explore in GA4.
• Create a blank free-form exploration.
• Add Hostname as a dimension and Event count (or Sessions) as a metric.
• Put Hostname in rows, and your metric in values.
• Review the list and note every hostname you want to keep, including the www and non-www versions, subdomains, and any payment or checkout domains.

How Do You Create a Hostname Filter in GA4?

You'll need Editor or Administrator access to the property. There's no regex support yet, so exact matching, adding one hostname at a time, gives you the most control.

• Go to Admin.
• Under Data collection and modification, select Data filters.
• Click Create filter.
• Choose Web hostname traffic.
• Name the filter (for example, "Approved Hostnames").
• Select Include only, or leave it on Exclude, depending on your approach.
• Pick a match type: Exactly matches, Begins with, Ends with, or Contains.
• Enter your approved or blocked hostnames.
• Set the filter state to Testing.
• Click Create.

How Do You Test a Filter Before Activating It?

Leave the filter in Testing for at least a few days, long enough to capture traffic across your main domain, subdomains, and anything else that legitimately sends data to the property. While testing, use the Test data filter name dimension in your reports to see exactly which events would have been blocked, without actually losing any data. Only flip it to Active once you're confident nothing important is getting caught.

Once active, the filter permanently affects incoming data. Blocked events aren't recoverable, and it does not affect data already collected.

What Mistakes Should You Avoid?

With an Exclude filter: forgetting a hostname means unwanted traffic keeps flowing through untouched. If new spam domains keep appearing, switch to an Include filter instead.

With an Include filter: forgetting a legitimate hostname, a subdomain, a payment gateway, or a staging domain your web development team still uses means that traffic silently disappears once the filter goes active. Double-check your hostname list against real report data before switching from Testing to Active.

What Are the Limitations?

Does it block Measurement Protocol spam? No. This is the detail most guides gloss over: events sent via the Measurement Protocol — a common vector for the exact kind of ghost spam this filter is meant to solve — bypass the include hostname filter entirely. If that's your main spam source, you'll need to protect your API secrets and address it separately. Don't rely on the hostname filter alone.

Does it clean up historical data? No. The filter only affects events collected after you activate it. To exclude bad hostnames from past reports, you'll still need a segment, comparison, or exploration filter applied manually.

Under an Include filter, events with an empty hostname are automatically blocked. That's usually a good thing — a missing hostname often signals junk traffic — but it can also catch legitimate events that don't include the page_location parameter, so it's worth checking your implementation before activating.